Method and apparatus for providing artificial intelligence-based procedure recommendation and dedicated care service
An AI-based system optimizes treatment recommendations and communication for foreign medical tourists by analyzing personal information and matching interpreters, addressing language barriers and cost transparency issues in medical tourism.
Patent Information
- Application Number
- KR1020260027605
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2026-02-11
- Publication Date
- 2026-11-11
- Estimated Expiration
- 2046-02-11
AI Technical Summary
Existing medical tourism platforms fail to provide personalized treatment recommendations, accurate communication, and comprehensive care for foreign medical tourists due to language barriers, information asymmetry, and lack of predictive analytics, leading to misunderstandings, unexpected costs, and post-procedure complications.
An AI-based system that analyzes personal information of foreign medical tourists to recommend customized plastic surgery procedures, calculates additional procedure items and estimated costs, and matches dedicated interpreters for seamless care throughout the process.
Provides optimized treatment information, transparent cost estimation, and smooth communication, ensuring a safe and reliable medical tourism experience by overcoming language barriers and predicting additional procedures.
Smart Images

Figure 112026018247969-PAT00001_ABST
Abstract
Description
Technology Field
[0001] The present invention relates to a method and apparatus for providing AI-based procedure recommendations and dedicated care services, and more specifically, to a method and apparatus for providing AI-based procedure recommendations and dedicated care services that analyzes the personal information of foreign medical tourists to recommend customized plastic surgery procedures, calculates additional procedure items and estimated cost ranges based on past customer data, and matches a dedicated interpreter to care for the entire process before and after the procedure. Background Technology
[0002] With the recent rise in international interest in Korea's K-beauty and medical technology, the number of foreign medical tourists, including those from Southeast Asia, is surging. According to data from the Korea Tourism Organization, medical tourism spending by six Southeast Asian countries increased by an annual average of 322% from 5.1 billion won in 2021 to 68.1 billion won in 2023, with a significant increase in the number of foreign patients visiting Korea, particularly in the fields of plastic surgery and dermatology.
[0003] However, despite this quantitative growth, there are several structural problems in the foreign medical tourism market.
[0004] Due to language barriers, foreign patients sometimes undergo procedures without a sufficient understanding of the details, side effects, and aftercare. Even when interpretation services are provided, accurate communication regarding medical terminology is not achieved. Furthermore, misunderstandings arise, particularly when additional procedures are required, as the necessity and costs are not adequately explained.
[0005] Due to information asymmetry in medical data, patients find it difficult to determine which procedures are suitable for their condition and end up relying entirely on hospital recommendations. In this process, procedures unsuitable for the patient's skin type or structure may be recommended, which can lead to dissatisfaction with the treatment results.
[0006] Lack of price transparency is the biggest cause of disputes in medical tourism for foreigners. Discrepancies frequently occur between initial estimates and final payment amounts, and disputes arise when unexpected additional costs are charged during procedures. In fact, the number of cases involving illegal brokers targeting foreign patients increased from 13 in 2020 to 67 in 2023, with a significant number of these related to overcharging.
[0007] There are also difficulties in post-procedure care. Since many foreign patients return to their home countries after the procedure, it is difficult for them to receive proper management for side effects or complications, and they often miss the timing for necessary follow-up visits. From the hospital's perspective, it is also difficult to track and manage patients who have returned home.
[0008] From the hospital's perspective, there are also difficulties such as communication issues with foreign patients, the complexity of appointment management, and the burden of high marketing costs. In particular, attracting patients through brokers requires paying a significant portion of the procedure cost as commissions and entails issues regarding the accuracy of patient information.
[0009] Existing medical tourism platforms have had limitations in addressing these issues. Most platforms are limited to simply listing hospital information or comparing prices, lacking personalized treatment recommendation features that consider the individual characteristics of users. The displayed prices are for basic procedures only, and predictive information regarding potential additional procedures or extra costs is not provided.
[0010] Even when some platforms provide interpretation services, they are often limited to one-time consultations, failing to provide comprehensive care throughout the entire process before and after the procedure. From a hospital management perspective, inefficiencies arise due to a lack of functions to systematically manage foreign patients' appointment information and procedure history.
[0011] Furthermore, existing systems lack predictive analytics capabilities utilizing historical customer data, failing to provide insights such as how frequently additional procedures occur in specific treatments or what patterns emerge within particular customer groups.
[0012] Therefore, there is a need for an integrated medical tourism platform that analyzes the individual characteristics of foreign medical tourists to recommend optimal procedures, transparently provides the likelihood of additional procedures and estimated costs based on past data, cares for the entire process before and after the procedure through dedicated interpreters, and enables the system to learn autonomously by receiving feedback on actual procedure results. The problem to be solved
[0013] The present invention was devised to solve the aforementioned problems, and aims to provide a method and apparatus for providing an AI-based procedure recommendation and dedicated care service that extracts similar customer groups based on the basic and medical information of foreign medical tourists to recommend customized plastic surgery procedures, analyzes the difference between the initial estimate and the final bill from past customer data to provide additional possible procedure items and an estimated cost range, classifies users into detailed customer groups through self-diagnosis questions to improve prediction accuracy, and matches a dedicated interpreter to care for the entire process before and after the procedure. means of solving the problem
[0014] A method for providing an artificial intelligence-based procedure recommendation and dedicated care service according to an embodiment of the present invention for achieving the above objective may include: receiving, from a user terminal, basic information including at least one of the user's nationality, currency, and preferred language, and medical information including at least one of age, skin condition, past procedure history, and allergy information; calculating the similarity between the basic information and medical information among past customer data to extract a customer group that is greater than or equal to a preset similarity threshold value, and determining a plastic surgery procedure suitable for the user based on the procedure history of the customer group; calculating an estimated cost range based on price information regarding the plastic surgery procedure among hospital data; matching a dedicated interpreter corresponding to the user's preferred language among a plurality of dedicated interpreters based on the basic information; and transmitting recommendation information including the plastic surgery procedure, the estimated cost range, and the matched dedicated interpreter to the user terminal.
[0015] Additionally, the step of calculating the estimated cost range may include: a step of identifying additional procedure items by comparing the items included in the initial estimate and the items included in the final invoice of customers who received the plastic surgery procedure among past customer data; a step of calculating the frequency of occurrence and the average additional cost for each additional procedure item; and a step of calculating the estimated cost range by reflecting the average additional cost of additional procedure items whose frequency of occurrence is above a preset threshold.
[0016] Additionally, the step of identifying the additionally generated procedure item may include receiving response data for a self-diagnosis question item generated in response to the plastic surgery procedure from the user terminal, subdividing the customer group based on the response data to classify the detailed customer group to which the user belongs, and identifying the history of additional procedures for customers belonging to the detailed customer group among past customer data.
[0017] In addition, the self-diagnosis question items may be characterized by being generated by extracting medical symptom information of customers who underwent additional procedures regarding the plastic surgery procedure from past customer data, and automatically converting the extracted medical symptom information into a question format.
[0018] Additionally, the method may further include the step of receiving procedure result information from a hospital system, including additional procedure items actually performed for the user and the final payment amount, and the step of transmitting post-procedure care information to the user terminal, including precautions for each recovery stage of the plastic surgery procedure, recommended management methods, and recommended timing for revisit.
[0019] Additionally, the step of receiving the procedure result information may include the step of storing the actual additional procedure item in past customer data in conjunction with the response data and the detailed customer group information, the step of calculating the prediction accuracy of the occurrence of additional procedure for the detailed customer group, and the step of regenerating the self-diagnosis question item if the prediction accuracy is less than a preset threshold.
[0020] Additionally, the step of transmitting the above-mentioned post-management information may include determining whether the actual additional procedure item that occurred is included in the procedure item identified in the additional procedure occurrence history of the detailed customer group, and transmitting the determination result to the user terminal.
[0021] Additionally, the step of transmitting the recommendation information may include: a step of calculating a price transparency score based on the number of additional procedure items and average additional costs for each of the multiple hospitals among the hospital data, where the frequency of occurrence is above a preset threshold; and a step of transmitting recommendation information about the hospitals to the user terminal in order of highest price transparency score.
[0022] Additionally, the method further includes the step of generating a customer management message based on the user's reservation information and treatment history, and transmitting the customer management message to the user terminal. The customer management message may include at least one of an automatic notification message generated at a time set prior to the reservation schedule based on the reservation information, a revisit guidance message transmitted at a recommended revisit time automatically calculated by analyzing the treatment history, and a customized treatment promotion generated by analyzing treatment information preferred by other users of the same nationality as the user from past customer data.
[0023] An AI-based procedure recommendation and dedicated care service provision device according to another embodiment of the present invention for achieving the above objective comprises a memory on which at least one program is recorded and a processor that executes said program, and said program may include instructions for performing the steps of: receiving from a user terminal basic information including at least one of the user's nationality, currency, and preferred language, and medical information including at least one of age, skin condition, past procedure history, and allergy information; calculating the similarity between said basic information and medical information among past customer data to extract a customer group that is greater than or equal to a preset similarity threshold value, and determining a plastic surgery procedure suitable for said user based on the procedure history of said customer group; calculating an estimated cost range based on price information for said plastic surgery procedure among hospital data; matching a dedicated interpreter corresponding to said user's preferred language among a plurality of dedicated interpreters based on said basic information; and transmitting recommendation information including said plastic surgery procedure, said estimated cost range, and said matched dedicated interpreter to said user terminal. Effects of the invention
[0024] According to the present invention as described above, by extracting similar customer groups based on personal information such as the nationality, age, skin condition, and past treatment history of foreign medical tourists, and analyzing the treatment history of the group to recommend customized plastic surgery procedures, it is possible to provide treatment information optimized for each individual user.
[0025] In addition, according to the present invention, by analyzing the difference between the initial estimate and the final bill in past customer data to calculate the likelihood of additional procedures and the average additional cost, the estimated cost range can be transparently provided to the user, and disputes arising from unexpected additional costs can be prevented in advance.
[0026] In addition, according to the present invention, a dedicated interpreter corresponding to the user's preferred language is automatically matched to overcome language barriers and improve the quality of medical services through smooth communication before and after the procedure.
[0027] In addition, according to the present invention, by integrating and providing recommendations for plastic surgery procedures, estimated cost ranges, and dedicated interpreter information, it is possible to support foreign medical tourists in having a safe and reliable medical tourism experience. Brief explanation of the drawing
[0028] FIG. 1 is a diagram showing the configuration of an artificial intelligence-based procedure recommendation and dedicated care service provision system according to one embodiment of the present invention. FIG. 2 is an overall flowchart of an artificial intelligence-based procedure recommendation and dedicated care service provision method according to one embodiment of the present invention. FIG. 3 is a flowchart illustrating a method for calculating an estimated cost range according to an embodiment of the present invention. FIG. 4 is a flowchart illustrating a method for identifying additional procedure items according to one embodiment of the present invention. FIG. 5 is a flowchart illustrating a post-management and feedback method according to an embodiment of the present invention. FIG. 6 is a flowchart illustrating a customer management message transmission method according to an embodiment of the present invention. FIG. 7 is a diagram showing a computer system according to one embodiment of the present invention. Specific details for implementing the invention
[0029] The advantages and features of the present invention and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the present invention, and the present invention is defined only by the scope of the claims.
[0030] The terms used in this specification are for describing embodiments and are not intended to limit the invention. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. The terms "comprises" and / or "comprising" used in this specification do not exclude the presence or addition of one or more other components in addition to the components mentioned. Throughout the specification, the same reference numerals refer to the same components, and "and / or" includes each of the mentioned components and all combinations of one or more. Although terms such as "first," "second," etc., are used to describe various components, these components are not limited by these terms. These terms are used merely to distinguish one component from another. Therefore, the first component mentioned below may be the second component within the technical scope of the invention.
[0031] Unless otherwise defined, all terms used herein (including technical and scientific terms) may be used in a meaning commonly understood by those skilled in the art to which the present invention pertains. Additionally, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.
[0032] Spatially relative terms such as "below," "beneath," "lower," "above," and "upper" may be used to facilitate the description of the relationship between one component and other components as illustrated in the drawings. Spatially relative terms should be understood as encompassing different orientations of components during use or operation, in addition to the orientations depicted in the drawings. For example, if a component depicted in a drawing is inverted, a component described as "below" or "beneath" of another component may be placed "above" of that component. Therefore, the exemplary term "below" may encompass both the lower and upper directions. Components may also be oriented in other directions, and accordingly, spatially relative terms may be interpreted according to the orientation.
[0033] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.
[0034] FIG. 1 is a diagram showing an artificial intelligence-based procedure recommendation and dedicated care service provision system according to one embodiment of the present invention.
[0035] Referring to FIG. 1, an artificial intelligence-based procedure recommendation and dedicated care service provision system according to one embodiment of the present invention includes a server (110), a user terminal (120), and a hospital system (130), and can operate by interoperating through a network.
[0036] The server (110) is a core component that provides AI-based procedure recommendations and dedicated care services, and can perform functions such as receiving user information from a user terminal (120), analyzing past customer data to recommend customized plastic surgery procedures, calculating an estimated cost range, and matching a dedicated interpreter.
[0037] As an optional embodiment, the server (110) can be implemented in various forms of computing infrastructure, such as a cloud environment, an on-premises server, a hybrid cloud, or an edge computing node.
[0038] As an optional embodiment, the server (110) may receive user's basic information and medical information from the user terminal (120). The basic information may include at least one of nationality, currency, and preferred language, and the medical information may include at least one of age, skin condition, history of past procedures, and allergy information.
[0039] As an optional embodiment, the server (110) may store and analyze past customer data. The past customer data may include basic information, medical information, procedure history, initial quotes, final invoices, etc. of customers who have previously used the service.
[0040] As an optional embodiment, the server (110) may extract similar customer groups by calculating the similarity between the received user information and past customer data. The similarity may be calculated by comprehensively considering multiple attributes such as nationality, age, skin condition, and past treatment history.
[0041] As an optional embodiment, the server (110) can determine a plastic surgery procedure suitable for the user by analyzing the procedure history of an extracted similar customer group. When determining the procedure, procedures with high frequency and high satisfaction within the group may be considered first.
[0042] As an optional embodiment, the server (110) may obtain price information for a procedure determined from hospital data, identify additional procedure items by comparing an initial estimate and a final invoice in past customer data, and determine an estimated cost range by calculating the frequency of occurrence and average additional cost of additional procedure items.
[0043] As an optional embodiment, the server (110) may match a dedicated interpreter corresponding to the user's preferred language from among a plurality of dedicated interpreters. When matching an interpreter, not only language ability but also expertise in the relevant field of treatment and past interpretation satisfaction may be considered.
[0044] As an optional embodiment, the server (110) can receive procedure result information from the hospital system (130), store it in past customer data, and continuously improve prediction accuracy.
[0045] The user terminal (120) is a device for using AI-based treatment recommendation and dedicated care services by communicating with the server (110) through a network, and can input user information and receive and display recommendation information from the server (110).
[0046] As an optional embodiment, the user terminal (120) can be implemented as various types of electronic devices such as a smartphone, tablet PC, laptop, or desktop computer.
[0047] As an optional embodiment, the user terminal (120) may receive nationality, age, skin condition, past treatment history, allergy information, etc. from the user and transmit it to the server (110).
[0048] As an optional embodiment, the user terminal (120) may display recommendation information received from the server (110) on a screen. The recommendation information may include recommended plastic surgery procedures, an estimated cost range, and information on a matched dedicated interpreter.
[0049] As an optional embodiment, the user terminal (120) may receive and display self-diagnosis question items from the server (110), and receive response data from the user and transmit it to the server (110).
[0050] As an optional embodiment, the user terminal (120) may receive after-care information, customer management messages, etc. from the server (110) and provide them to the user.
[0051] The hospital system (130) is a system equipped in a plastic surgery or dermatology hospital and can transmit and receive hospital information, price information, reservation information, procedure result information, etc. by linking with the server (110).
[0052] As an optional embodiment, the hospital system (130) may be implemented in the form of a terminal, kiosk, POS system, etc., provided in the hospital.
[0053] As an optional embodiment, the hospital system (130) may provide the server (110) with a list of procedures provided by the hospital and price information for each procedure.
[0054] As an optional embodiment, the hospital system (130) can receive reservation information from the server (110) and reflect it in the hospital's reservation schedule.
[0055] As an optional embodiment, the hospital system (130) may transmit procedure result information, including additional procedure items and final payment amount that actually occurred after the procedure is completed, to the server (110).
[0056] As an optional embodiment, a plurality of hospital systems (130) may be linked with a server (110), and the server (110) may collect information on each of the plurality of hospitals and recommend the optimal hospital to the user.
[0057] As an optional embodiment, the server (110) may include an artificial intelligence model. The artificial intelligence model may perform functions such as calculating the similarity between user information and past customer data, recommending procedures, and predicting the probability of additional procedures occurring.
[0058] As an optional embodiment, the server (110) may convert the user's nationality, age, skin condition, past treatment history, etc., into numeric vectors and use them as inputs to an artificial intelligence model. Nationality may be converted using one-hot encoding, age using normalization, and past treatment history using multi-hot encoding.
[0059] As an optional embodiment, the artificial intelligence model may be implemented as at least one of a deep neural network, a random forest, gradient boosting, and logistic regression.
[0060] As an optional embodiment, the server (110) may train an artificial intelligence model using past customer data as training data. The training data may consist of pairs of (user information, procedure history, whether additional procedures occurred).
[0061] As an optional embodiment, the server (110) may add the procedure result information to the training data whenever it receives it and periodically retrain the artificial intelligence model to improve prediction accuracy.
[0062] As an optional embodiment, when calculating similarity, the server (110) may use cosine similarity, Euclidean distance, or distance in a learned embedding space.
[0063] FIG. 2 is an overall flowchart of an artificial intelligence-based procedure recommendation and dedicated care service provision method according to one embodiment of the present invention.
[0064] Referring to FIG. 2, in one embodiment, the server (110) can first receive user basic information and medical information from the user terminal (120) (S210).
[0065] As an optional embodiment, basic information may include at least one of the user's nationality, currency, and preferred language. Nationality information indicates the country to which the user belongs and can be used to analyze the preferred treatment patterns of users of the same nationality. Currency information indicates the currency to be used for price display and payment. Preferred language information indicates the language the user intends to use for communication and can serve as a criterion for interpreter matching.
[0066] As an optional embodiment, medical information may include at least one of age, skin condition, history of past procedures, and allergy information. Age information indicates the user's age and can be used to recommend appropriate procedures for each age group. Skin condition information may include skin type, skin tone, and skin problems (acne, wrinkles, pigmentation, etc.). History of past procedures may include the type and timing of cosmetic or skin procedures previously received by the user. Allergy information may include whether there is an allergy to specific ingredients or drugs.
[0067] As an optional embodiment, the user terminal (120) may receive the information from the user through a user interface. The user interface may include a text input field, a dropdown menu, a checkbox, an image upload function, etc.
[0068] Next, the server (110) can determine a suitable plastic surgery procedure for the user (S220).
[0069] As an optional embodiment, the server (110) may calculate the similarity between the basic information and medical information among past customer data. When calculating the similarity, the similarity for multiple attributes such as nationality, age, skin condition, and past procedure history may be calculated separately, and the weighted average thereof may be used to calculate the overall similarity.
[0070] As an optional embodiment, the server (110) may form a customer group by extracting customers whose overall similarity is greater than or equal to a preset similarity threshold value. The similarity threshold value may be set, for example, to 70% or higher.
[0071] As an optional embodiment, the server (110) can analyze the treatment history of the extracted customer group to determine a suitable cosmetic procedure for the user. When determining the procedure, procedures that are frequently performed in the group may be given priority. For example, if double eyelid surgery is performed most frequently in a group of Thai women in their late 20s, double eyelid surgery may be recommended to a new user with the same characteristics.
[0072] As an optional embodiment, the server (110) may determine the procedure by considering the procedure satisfaction, procedure result evaluation, etc. Procedures with low satisfaction, even if frequent, may be excluded from the recommendation.
[0073] Next, the server (110) can calculate the estimated cost range of the plastic surgery procedure (S230).
[0074] As an optional embodiment, the server (110) may obtain price information for a plastic surgery procedure determined from hospital data. The price information may include basic procedure costs.
[0075] As an optional embodiment, the server (110) can identify additional procedure items by analyzing the initial estimate and final invoice of customers who received the same plastic surgery procedure in past customer data.
[0076] As an optional embodiment, the server (110) may calculate an estimated cost range by reflecting the frequency of occurrence of additional procedure items and the average additional cost. The estimated cost range may include a minimum estimated cost (basic procedure cost) and a maximum estimated cost (basic procedure cost + additional procedure cost).
[0077] Next, the server (110) can match a dedicated interpreter corresponding to the user's preferred language (S240).
[0078] As an optional embodiment, the server (110) may select an interpreter capable of speaking the user's preferred language from among a plurality of dedicated interpreters. For example, if the user's preferred language is Thai, interpreters capable of speaking Thai may be selected as candidates.
[0079] As an optional embodiment, the server (110) may match a final dedicated interpreter from among the selected interpreters by considering their expertise in the relevant field of treatment, past interpretation satisfaction, available schedule, etc.
[0080] As an optional embodiment, a matched dedicated interpreter can care for the user throughout the entire process, including pre-procedure consultation, interpretation on the day of the procedure, and guidance on post-procedure care.
[0081] Next, the server (110) can transmit recommendation information to the user terminal (120) (S250).
[0082] As an optional embodiment, the recommendation information may include recommended cosmetic procedures, an estimated cost range, and information on a matched dedicated interpreter.
[0083] As an optional embodiment, recommended cosmetic procedure information may include the name of the procedure, a description of the procedure, the effects of the procedure, the estimated time required, the recovery period, etc.
[0084] As an optional embodiment, the estimated cost range includes a minimum estimated cost and a maximum estimated cost, and may also provide a description of additional possible procedure items.
[0085] As an optional embodiment, dedicated interpreter information may include the interpreter's name, photo, language proficiency, field of expertise, past customer reviews, etc.
[0086] As an optional embodiment, the user terminal (120) may display the received recommendation information on a screen and receive a reservation request from the user. When the user approves the recommendation information and requests a reservation, the server (110) may transmit the reservation information to the hospital system (130) and the dedicated interpreter to confirm the reservation.
[0087] FIG. 3 is a flowchart illustrating a method for calculating an estimated cost range according to an embodiment of the present invention.
[0088] Referring to FIG. 3, FIG. 3 details step S230 of FIG. 2 and illustrates a process of analyzing past customer data to predict the possibility of additional procedures and calculating an estimated cost range.
[0089] In one embodiment, the server (110) can identify additional occurrence procedure items (S310).
[0090] As an optional embodiment, the server (110) may retrieve the initial estimate and final invoice of customers who received the plastic surgery procedure from past customer data. The initial estimate includes the estimated procedure items and costs presented by the hospital during the pre-procedure consultation phase, and the final invoice includes the procedure items actually incurred and the final payment amount after the completion of the procedure.
[0091] As an optional embodiment, the server (110) can compare items included in the initial estimate with items included in the final invoice and identify items that are in the final invoice but not in the initial estimate as additional procedure items.
[0092] As an optional embodiment, for example, if the initial estimate for double eyelid surgery included only the item "incisional double eyelid surgery," but the final invoice includes "upper eyelid lift" and "eyelid correction" in addition to "incisional double eyelid surgery," then "upper eyelid lift" and "eyelid correction" can be identified as additional procedure items.
[0093] Next, the server (110) can calculate the frequency of occurrence for each procedure item (S320).
[0094] As an optional embodiment, the server (110) may calculate the frequency at which an item was added in past customer data for each identified additional procedure item. The frequency of occurrence may be calculated as the ratio of customers for whom the additional procedure occurred out of the total number of customers.
[0095] As an optional embodiment, for example, if "upper eyelid lifting" is added to 65 of 100 past customers who have undergone double eyelid surgery, the incidence of "upper eyelid lifting" can be calculated as 65%.
[0096] As an optional embodiment, the server (110) may also calculate the average additional cost for each additional procedure item. The average additional cost may be calculated as the average of the additional costs incurred by customers who received the additional procedure.
[0097] Next, the server (110) can determine whether the occurrence frequency is greater than or equal to a preset threshold (S330).
[0098] As an optional embodiment, the threshold can be set to, for example, 30%. The threshold can be adjusted by a system administrator or a hospital operator.
[0099] As an optional embodiment, additional procedure items with a frequency of occurrence above a threshold are judged to be items with a high probability of occurrence and are reflected in the estimated cost, while items below the threshold are judged to be items with a low probability of occurrence and are not reflected in the estimated cost.
[0100] If the frequency of occurrence is greater than a threshold (e.g.), the server (110) can reflect the average additional cost (S340).
[0101] As an optional embodiment, the server (110) can calculate the maximum estimated cost by adding the average additional cost of additional procedure items whose occurrence frequency is above a threshold to the basic procedure cost.
[0102] As an optional embodiment, for example, if the basic cost of double eyelid surgery is 2 million won, the average additional cost of "upper eyelid lift" with a frequency of 65% is 800,000 won, and the average additional cost of "eyelid correction" with a frequency of 40% is 500,000 won, the maximum estimated cost can be calculated as 2 million won + 800,000 won + 500,000 won = 3.3 million won.
[0103] As an optional embodiment, the server (110) can increase transparency by providing the user with additional possible procedure item names, frequency of occurrence, and average additional cost.
[0104] If the frequency of occurrence is below the threshold (no), the server (110) may not reflect the average additional cost (S341).
[0105] As an optional example, additional procedure items with a frequency below a threshold are not included in the calculation of estimated costs, but may be separately notified to the user as "rarely occurring additional procedures."
[0106] As an optional embodiment, the server (110) may provide the user with an estimated cost range in the form of [minimum estimated cost, maximum estimated cost], where the basic procedure cost is the minimum estimated cost and the amount obtained by adding additional procedure costs to the basic procedure cost, where the frequency of occurrence is greater than or equal to a threshold, is the maximum estimated cost.
[0107] FIG. 4 is a flowchart illustrating a method for identifying additional procedure items according to one embodiment of the present invention.
[0108] Referring to Fig. 4, Fig. 4 details step S310 of Fig. 3 and illustrates a process of classifying users into detailed customer groups through self-diagnosis questions to increase the accuracy of additional procedure predictions.
[0109] In one embodiment, the server (110) can receive self-diagnosis response data (S410).
[0110] As an optional embodiment, the server (110) may transmit self-diagnosis question items generated in response to a plastic surgery procedure to the user terminal (120). The self-diagnosis question items may include questions to check the user's symptoms, condition, preferences, etc. related to the procedure.
[0111] As an optional embodiment, self-diagnosis questions regarding double eyelid surgery, such as “Do you feel your eyelids are drooping?”, “Do you feel tension in your forehead when opening your eyes?”, and “Do you have epicanthal folds?” may be provided.
[0112] As an optional embodiment, the user terminal (120) may receive a response to each question from the user and transmit it to the server (110). The response may be collected in various forms, such as yes / no, multiple choice, or a 1-5 point scale.
[0113] As an optional embodiment, self-diagnosis question items may be generated by extracting medical symptom information from past customer data regarding customers who underwent additional cosmetic procedures, and automatically converting the extracted medical symptom information into a question format. For example, if "ptosis (drooping eyelids)" is a common symptom among customers who underwent "upper eyelid lift" after double eyelid surgery, this can be automatically converted into the question, "Do you feel like your eyelids are drooping?"
[0114] Next, the server (110) can classify detailed customer groups (S420).
[0115] As an optional embodiment, the server (110) may further subdivide the previously extracted customer group based on the received response data. Subdivision may be performed by analyzing the similarity of response patterns.
[0116] As an optional embodiment, for example, within a group of Thai women in their late 20s who want double eyelid surgery, it can be subdivided into "subgroup A with symptoms of eyelid ptosis and epicanthal folds" and "subgroup B with no symptoms of eyelid ptosis but dark circles."
[0117] As an optional embodiment, the server (110) can calculate a similarity by comparing the user's response data with the response data of past customers and classify the user into the detailed customer group with the highest similarity.
[0118] As an optional embodiment, various machine learning techniques, such as clustering algorithms and classification algorithms, can be utilized for detailed group classification.
[0119] Next, the server (110) can identify the history of additional procedures occurring within the same detailed customer group (S430).
[0120] As an optional embodiment, the server (110) can query the history of additional procedures for customers belonging to a detailed customer group classified by user among past customer data.
[0121] As an optional embodiment, for example, if a user is classified as "Subgroup A with symptoms of eyelid ptosis and epicanthal folds," it is possible to analyze the additional procedure history of past customers belonging to the same subgroup to identify that "upper eyelid lifting" occurs with an 80% frequency and "epicanthoplasty" occurs with a 55% frequency.
[0122] As an optional embodiment, the frequency of additional procedures by subgroup may be more accurate than the average frequency of the entire customer group. This is because the analysis is conducted on customers with the same symptoms and characteristics.
[0123] As an optional embodiment, the server (110) can calculate an estimated cost range by performing the process after step S320 of FIG. 3 based on the history of additional procedures of the identified subgroup. In this case, the accuracy of the prediction can be improved because data from the subgroup of customers, rather than the entire customer group, is used.
[0124] As an optional embodiment, detailed group classification through self-assessment questions may be performed optionally. If the user skips the self-assessment, the estimated cost may be calculated based on the average data of the entire customer group.
[0125] FIG. 5 is a flowchart illustrating a post-management and feedback method according to an embodiment of the present invention.
[0126] Referring to Fig. 5, Fig. 5 illustrates a process of receiving procedure result information after the completion of the procedure, providing follow-up care information, and continuously improving the prediction accuracy of the system by feeding back the actual procedure results to past customer data.
[0127] In one embodiment, the server (110) can receive procedure result information from the hospital system (130) (S510).
[0128] As an optional embodiment, the procedure result information may include additional procedure items actually performed and the final payment amount. Additional procedure items actually performed refer to procedures that were not included in the initial estimate but were additionally conducted during the procedure.
[0129] As an optional embodiment, for example, if a user who has booked double eyelid surgery additionally receives "upper eyelid lifting" and "eye shape correction" during the procedure, the hospital system (130) can transmit these additional procedure items and the final payment amount to the server (110).
[0130] As an optional embodiment, the server (110) may store the received procedure result information in historical customer data in conjunction with the user's response data and detailed customer group information. Through this, the correlation between the pre-procedure self-diagnosis response and the actual occurrence of additional procedures can be accumulated as data.
[0131] As an optional embodiment, the server (110) can calculate the accuracy of predicting the occurrence of additional procedures for a detailed customer group. The accuracy of the prediction can be calculated as the rate of agreement between the additional procedure items predicted by the server (110) and the additional procedure items that actually occurred.
[0132] As an optional embodiment, for example, if the server (110) predicts that "upper eyelid lifting" and "epicanthoplasty" are additional possible procedures, but in reality "upper eyelid lifting" and "eye shape correction" occur, the prediction accuracy can be calculated as 50% because one of the two matches.
[0133] As an optional embodiment, the server (110) may regenerate self-diagnosis question items for the relevant detailed customer group if the prediction accuracy is below a preset threshold (e.g., 70%). Since low prediction accuracy means that the current self-diagnosis question is not suitable for predicting the occurrence of additional procedures, it needs to be improved into a more appropriate question.
[0134] As an optional embodiment, when regenerating self-diagnosis questions, the server (110) may reanalyze recently accumulated procedure result data to newly extract common symptoms or characteristics of customers who have undergone additional procedures and convert them into questions. For example, new questions such as "Do wrinkles form on the forehead when you open your eyes?" which were not captured in existing questions may be added.
[0135] Next, the server (110) can transmit post-management information to the user terminal (120) (S520).
[0136] As an optional embodiment, post-operative care information may include precautions for each recovery stage of the cosmetic procedure, recommended care methods, and recommended timing for revisit.
[0137] As an optional embodiment, precautions for each recovery stage may include precautions to be taken at each stage, such as immediately after the procedure, day 1, day 3, day 7, and day 14. For example, information such as "cold compress for 48 hours after the procedure," "no alcohol consumption for 7 days after the procedure," and "removal of stitches 14 days after the procedure" may be provided.
[0138] As an optional embodiment, the recommended care method may include a facial cleansing method, a guide to using cosmetics, and a method for managing the treatment area, such as UV protection.
[0139] As an optional example, the recommended time for a follow-up visit may vary depending on the type of procedure. For example, a follow-up procedure for Botox may be recommended after 4-6 months, and a follow-up procedure for fillers may be recommended after 6-12 months.
[0140] As an optional embodiment, the server (110) can determine whether an additional procedure item that actually occurred is included in the procedure item identified in the additional procedure occurrence history of a detailed customer group, and transmit the determination result to the user terminal (120).
[0141] As an optional embodiment, if the actual additional procedure performed was within the pre-predicted range, a message stating "The additional cost incurred in this procedure is within the pre-notified estimated range" may be provided. This allows the user to confirm that the system's prediction was accurate and to increase confidence.
[0142] As an optional embodiment, if an additional procedure actually performed falls outside the predicted range, a message stating "An unexpected additional procedure has occurred" may be provided, along with an explanation from the hospital regarding the reason why the procedure was necessary. This helps the user understand the situation and minimizes dissatisfaction.
[0143] As an optional embodiment, the server (110) may provide follow-up information in the user's native language through a dedicated interpreter. The interpreter may answer the user's questions during the follow-up stage and mediate communication with the hospital if necessary.
[0144] As an optional embodiment, the server (110) may send automatic notification messages at each recovery stage. For example, on the 7th day after the procedure, a notification saying "Please check the stitch removal schedule" may be automatically sent.
[0145] FIG. 6 is a flowchart illustrating a customer management message transmission method according to an embodiment of the present invention.
[0146] Referring to FIG. 6, FIG. 6 illustrates the process of generating and transmitting customer management messages, such as automatic notification messages, revisit guidance messages, and customized treatment promotions, based on the user's reservation information and treatment history.
[0147] In one embodiment, the server (110) can send a customer management message to the user terminal (120) (S610).
[0148] As an optional embodiment, the customer management message may include at least one of an automatic notification message, a return visit guidance message, and a customized treatment promotion.
[0149] As an optional embodiment, automatic notification messages may be generated at a set time prior to the scheduled appointment based on the user's reservation information. For example, a reservation confirmation and preparation guidance message may be automatically sent one day before the scheduled procedure date, two hours before the procedure date, etc.
[0150] As an optional embodiment, the automated notification message may include the appointment date and time, hospital location, dedicated interpreter information, and pre-procedure precautions (e.g., fasting, prohibition of makeup, etc.). The message may be provided in the user's preferred language.
[0151] As an optional embodiment, the revisit guidance message may be sent at a recommended revisit time automatically calculated by analyzing the user's treatment history. The server (110) stores the duration of the effect for each type of treatment in a database and can automatically send a revisit guidance when the corresponding period has elapsed since the date of the treatment.
[0152] As an optional embodiment, for example, a message saying "The duration of the Botox effect is about to expire. Would you like to schedule a re-treatment?" may be sent to a user who has received a Botox treatment at the fourth month after the treatment.
[0153] As an optional embodiment, the recommended timing for a return visit may be applied differently for each individual. The server (110) can learn the time when a user actually returned after a past procedure and calculate the user's personalized return visit cycle. For example, while the cycle for Botox is generally 4-6 months, if a specific user has a history of always returning after 5 months, a notification message can be sent to that user at the 5th month.
[0154] As an optional embodiment, a customized treatment promotion may be generated by analyzing treatment information preferred by other users of the same nationality as the user from past customer data. The server (110) may recommend treatments that the user may be interested in by considering the user's nationality, age group, past treatment history, etc.
[0155] As an optional embodiment, for example, a Thai female user who has undergone double eyelid surgery may be guided to information on procedures such as "rhinoplasty" and "lifting" that are frequently received by other users of the same nationality, as well as discount events.
[0156] As an optional embodiment, when creating a customized treatment promotion, the server (110) can analyze past customer data to see what additional treatments customers with a profile similar to the user have received after the first treatment. For example, if 40% of Thai women in their late 20s who have had double eyelid surgery have received additional lifting treatments within 6 months, the lifting treatment can be recommended to new users with the same characteristics.
[0157] As an optional embodiment, customer management messages may be sent through various channels, such as push notifications, SMS, email, and in-app messages. Users can set their preferred notification channel.
[0158] As an optional embodiment, the server (110) may adjust the frequency of sending customer management messages. Since sending excessive messages may cause inconvenience to the user, messages may be selected and sent according to importance. For example, reservation notifications may be classified as mandatory messages and must be sent, while promotions may be classified as optional messages and sent only when the user has consented to receive them.
[0159] As an optional embodiment, the server (110) can optimize the message sending strategy by analyzing the user's message response (view rate, click rate, scheduled conversion rate, etc.). By learning message types with good responses and time zones for sending, the effectiveness of future messages can be enhanced.
[0160] As an optional embodiment, customer care messages may be sent via a dedicated interpreter. In particular, regarding follow-up visit notices or important aftercare information, the interpreter can contact the user directly to provide detailed explanations and answer questions. This allows for the provision of a higher quality of care than simple automated messages.
[0161] As an optional embodiment, the hospital can manage the appointment status, procedure history, and revisit schedule of foreign patients through the server (110), thereby reducing the burden of managing foreign patients and improving marketing efficiency.
[0162] FIG. 7 is a diagram showing a computer system according to one embodiment of the present invention.
[0163] A method for providing artificial intelligence-based treatment recommendations and dedicated care services according to one embodiment of the present invention can be implemented in a computer system (1000), such as a computer-readable recording medium.
[0164] Referring to FIG. 7, a computer system (1000) may include one or more processors (1010), memory (1030), user interface input device (1040), user interface output device (1050), and storage (1060) that communicate with each other via a bus (1020). Additionally, the computer system (1000) may further include a network interface (1070) connected to a network (1080). The processor (1010) may be a central processing unit or a semiconductor device that executes processing instructions stored in memory (1030) or storage (1060). Memory (1030) and storage (1060) may be various forms of volatile or non-volatile storage media. For example, memory may include ROM (1031) or RAM (1032).
[0165] The specific embodiments described in this invention are examples and do not limit the scope of the invention in any way. For the sake of brevity of the specification, descriptions of prior electronic configurations, control systems, software, and other functional aspects of said systems may be omitted. Additionally, the connections of lines or connecting members between components shown in the drawings are illustrative of functional connections and / or physical or circuit connections, and may be replaced or additionally represented as various functional connections, physical connections, or circuit connections in actual devices. Furthermore, unless specifically stated as “essential,” “importantly,” etc., a component may not be strictly necessary for the application of the invention.
[0166] Accordingly, the scope of the present invention should not be limited to the embodiments described above, and all scopes equivalent to or equivalently modified from the claims set forth below, as well as the claims set forth below, shall be considered to fall within the scope of the concept of the present invention.
[0167] The steps of the method or algorithm described in connection with embodiments of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), Flash Memory, a hard disk, a removable disk, a CD-ROM, or any form of computer-readable recording medium well known in the art to which the present invention belongs.
[0168] Although embodiments of the present invention have been described above with reference to the attached drawings, those skilled in the art will understand that the present invention may be implemented in other specific forms without altering its technical concept or essential features. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. Explanation of the symbols
[0169] 110: Server 120: User terminal 130: Hospital System 1000: Computer System 1010: Processor 1020: Bus 1030: Memory 1031: ROM 1032: RAM 1040: User interface input device 1050: User interface output device 1060: Storage 1070: Network Interface 1080: Network
Claims
Claim 1 A method for providing AI-based procedure recommendations and dedicated care services performed by a server, comprising: receiving from a user terminal basic information including at least one of the user's nationality, currency, and preferred language, and medical information including at least one of age, skin condition, past procedure history, and allergy information; calculating the similarity between the basic information and medical information among past customer data to extract a customer group that is greater than or equal to a preset similarity threshold value, and determining a plastic surgery procedure suitable for the user based on the procedure history of the customer group; calculating an estimated cost range based on price information regarding the plastic surgery procedure among hospital data; matching a dedicated interpreter corresponding to the user's preferred language among a plurality of dedicated interpreters based on the basic information; transmitting recommendation information including the plastic surgery procedure, the estimated cost range, and the matched dedicated interpreter to the user terminal; and receiving procedure result information including additional procedure items actually performed for the user and the final payment amount from a hospital system. The method includes the step of transmitting post-care information to the user terminal, including precautions for each recovery stage of the plastic surgery procedure, recommended management methods, and recommended timing for revisiting. The step of calculating the estimated cost range comprises: receiving response data for self-diagnosis question items generated in response to the plastic surgery procedure from the user terminal; subdividing the customer group based on the response data to classify the detailed customer group to which the user belongs; comparing items included in the initial estimate and items included in the final invoice of customers who received the plastic surgery procedure among past customer data to identify additional procedure items that occurred from the history of additional procedures of customers belonging to the detailed customer group; and calculating the frequency of occurrence and average additional cost for each additional procedure item.The method includes a step of calculating the estimated cost range by reflecting the average additional cost of additional procedure items whose occurrence frequency exceeds a preset threshold, wherein the self-diagnosis question items are generated by extracting medical symptom information of customers who underwent additional procedures for the plastic surgery procedure from past customer data and automatically converting the extracted medical symptom information into a question format, and the step of transmitting the recommendation information includes: a step of calculating a price transparency score based on the number of additional procedure items whose occurrence frequency exceeds a preset threshold and the average additional cost for each of a plurality of hospitals among the hospital data; and a step of transmitting recommendation information about hospitals to the user terminal in order of the highest price transparency score, and the step of receiving the procedure result information includes: a step of storing the actual additional procedure items that occurred in past customer data in conjunction with the response data and the detailed customer group information; a step of calculating the prediction accuracy of additional procedure occurrence for the detailed customer group; and a step of regenerating the self-diagnosis question items if the prediction accuracy is less than a preset threshold. Method for providing AI-based procedure recommendations and dedicated care services. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 delete Claim 6 delete Claim 7 A method for providing AI-based treatment recommendation and dedicated care services, wherein, in claim 1, the step of transmitting the post-treatment information includes determining whether the actual additional treatment item that occurred is included in the treatment item identified in the additional treatment occurrence history of the detailed customer group, and transmitting the determination result to the user terminal. Claim 8 delete Claim 9 A method for providing AI-based treatment recommendation and dedicated care services, further comprising the step of generating a customer management message based on the user’s reservation information and treatment history among hospital data and transmitting the customer management message to the user terminal, wherein the customer management message comprises at least one of an automatic notification message generated at a time set prior to the reservation schedule based on the reservation information, a revisit guidance message transmitted at a recommended revisit time automatically calculated by analyzing the treatment history, and a customized treatment promotion generated by analyzing treatment information preferred by other users of the same nationality as the user from past customer data. Claim 10 A memory in which at least one program is recorded; and a processor for executing said program, wherein the program comprises: receiving from a user terminal basic information including at least one of the user’s nationality, currency, and preferred language, and medical information including at least one of age, skin condition, past procedure history, and allergy information; calculating the similarity between said basic information and medical information among past customer data to extract a customer group that is greater than or equal to a preset similarity threshold value, and determining a plastic surgery procedure suitable for said user based on the procedure history of said customer group; calculating an estimated cost range based on price information for said plastic surgery procedure among hospital data; matching a dedicated interpreter corresponding to said user’s preferred language among a plurality of dedicated interpreters based on said basic information; transmitting recommendation information including said plastic surgery procedure, said estimated cost range, and said matched dedicated interpreter to said user terminal; and receiving procedure result information including actual additional procedure items and a final payment amount for said user from a hospital system. The method includes commands for performing the step of transmitting post-treatment information to the user terminal, including precautions for each recovery stage of the plastic surgery procedure, recommended management methods, and recommended timing for revisiting. The step of calculating the estimated cost range comprises: receiving response data for self-diagnosis question items generated in response to the plastic surgery procedure from the user terminal; subdividing the customer group based on the response data to classify the detailed customer group to which the user belongs; comparing items included in the initial estimate and items included in the final invoice of customers who received the plastic surgery procedure among past customer data to identify additional procedure items that occurred from the history of additional procedures of customers belonging to the detailed customer group; and calculating the frequency of occurrence and average additional cost for each additional procedure item.The AI-based procedure recommendation and dedicated care service providing device includes the step of calculating the estimated cost range by reflecting the average additional cost of additional procedure items whose occurrence frequency is above a preset threshold, wherein the self-diagnosis question items are generated by extracting medical symptom information of customers who underwent additional procedures for the plastic surgery procedure from past customer data and automatically converting the extracted medical symptom information into a question form, and the step of transmitting the recommendation information includes the step of calculating a price transparency score based on the number of additional procedure items whose occurrence frequency is above a preset threshold and the average additional cost for each of a plurality of hospitals among the hospital data; and the step of transmitting recommendation information for hospitals to the user terminal in order of the highest price transparency score, and the step of receiving the procedure result information includes the step of storing the actual additional procedure items that occurred in past customer data in conjunction with the response data and the detailed customer group information; the step of calculating the prediction accuracy of additional procedure occurrence for the detailed customer group; and the step of regenerating the self-diagnosis question items when the prediction accuracy is below a preset threshold.
Citation Information
Patent Citations
Healthcare cost transparency systems and methods
US20120054119A1
Estimate Analysis and Validation
US20140095188A1
Method for Helping Patients Find Treatments Based on Similar Patients' Experiences
US20140244292A1
Adaptive patient questionnaire generation system and method
US20180330802A1
